Physically damage problematic plants: freshwater swamps
Overall effectiveness category Unknown effectiveness (limited evidence)
Number of studies: 2
Background information and definitions
Physical damage to plants may kill them directly, increase their susceptibility to disease, slow their growth and/or prevent reproduction. This section considers actions, other than cutting and mowing, that physically damage problematic plants. This includes crushing stems, removing seed heads and girdling trees (removing a strip of bark from the entire circumference of a trunk, stem or branch), as well as soil disturbance that damages the underground parts of problematic plants. Removing fragments of the damaged plants might be desirable to prevent regrowth.
For this action, “vegetation” refers to overall or non-target vegetation. Studies that only report responses of target problematic plants have not been summarized.
Related actions: Use cutting/mowing to control problematic herbaceous plants; Use cutting to control problematic trees/large shrubs; Remove surface soil/sediment; Bury surface soil/sediment or Disturb soil/sediment surface other than to control problematic plants.
Supporting evidence from individual studies
A controlled study in 2002–2004 aiming to restore a swamp in a reed canarygrass Phalaris arundinacea stand in Wisconsin, USA (Hovick & Reinartz 2007) found that ploughing after spraying herbicide increased plant diversity and richness more than spraying alone, but that ploughing had no additional effect on the number of tree seedlings. After two growing seasons, the vegetation was more diverse in a ploughed/sprayed plot than in plots that had only been sprayed (data reported as a diversity index). The same was true for overall plant richness (ploughed/sprayed: 11.3; sprayed: 6.6 species/m2). However, the treatments did not significantly differ in native plant richness (ploughed/sprayed: 5.5; sprayed: 4.0 species/m2), or in the density of non-planted tree seedlings (ploughed/sprayed: 56; sprayed: 25 seedlings/m2). The study also reported differences between treatments in the abundance of individual plant species (statistical significance not assessed). For example, common vervain Verbena hastata was more abundant in the ploughed/sprayed plot (100% of quadrats; 20% cover) than sprayed plots (40% of quadrats; 3% cover). Reed canarygrass was less abundant in the ploughed/sprayed plot (40% of quadrats; 18% cover) than sprayed plots (100% of quadrats; 73% cover). Methods: Nine plots were established in a canarygrass-invaded wetland. All nine plots were sprayed with herbicide (Roundup®) in November 2002, and planted with tree/shrub seedlings (roughly 1 seedling/m2) in spring 2003. One plot was also ploughed, before planting, in spring 2003. This plot was slightly higher and drier than the unploughed plots. In August 2004, plant species and their cover were surveyed in ten 1-m2 quadrats/treatment, ignoring planted trees/shrubs.Study and other actions tested
A replicated, controlled study in 2006–2009 in a floodplain swamp clearing invaded by reed canarygrass Phalaris arundinacea in Wisconsin, USA (Thomsen et al. 2012) found that cutting, disking and applying herbicide to invaded plots increased tree seedling abundance after 1–3 years, and increased cover of herbs other than canarygrass after three years. In three of three years following intervention, treated plots contained more tree seedlings (4–44 seedlings/m2) than untreated plots (0–5 seedlings/m2). At the same time, treated plots had lower reed canarygrass cover (7–31%) than untreated plots (83–92%). Cover of herbs other than reed canarygrass did not significantly differ between treated and untreated plots in the first two years after intervention (treated: 15–47%; untreated: 16–22%), but was higher in treated than untreated plots in the third year (treated: 35–58%; untreated: 12%). Methods: In November 2006, twenty plots (roughly 810 m2) were established in a storm-created clearing within a floodplain swamp. Sixteen canarygrass-dominated plots were treated by cutting the vegetation (with a mechanical mulcher), disking the soil, and applying herbicide (four combinations of herbicide type and dose; repeated applications in summer and autumn until November 2008). The other four plots received none of these interventions. The study does not distinguish between the effects of cutting, disking and applying herbicide. Some tree species were planted and/or sown across the whole clearing. Vegetation (excluding planted trees) was surveyed in August 2007–2009, in four 2.25-m2 quadrats/plot.Study and other actions tested